Detection of Chlamydophila pneumoniae IgG in paired serum samples: comparison of serological techniques in pneumonia cases

Transcription

1 APMIS 114: , 2006 Printed in Denmark. All rights reserved C 2006 The Authors Journal Compilation C 2006 APMIS ISSN Detection of Chlamydophila pneumoniae IgG in paired serum samples: comparison of serological techniques in pneumonia cases FERNANDO DE ORY, MARÍA EULALIA GUISASOLA and JOSÉ MARÍA EIROS Servicio de Microbiología Diagnóstica, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain De Ory F, Guisasola ME, Eiros JM. Detection of Chlamydophila pneumoniae IgG in paired serum samples: comparison of serological techniques in pneumonia cases. APMIS 2006;114: Serological diagnosis of Chamydophila pneumoniae is usually undertaken by complement fixation test (CFT) or by microimmunofluorescence (MIF). A number of commercial methods for detecting C. pneumoniae-specific IgG have been developed. The aim of this study was to compare the performance characteristics of six methods for the diagnosis of pneumonia due to C. pneumoniae, including CFT (in house), MIF (Vircell, Spain), and four ELISAs (Medac, Germany; Savyon, Israel; Serion, Germany; and DRG, Germany). ELISA-Medac, ELISA-Savyon, ELISA-DRG and MIF use C. pneumoniae antigens while ELISA-Serion and CFT use Chlamydophila genus-specific antigen. Acute and convalescent samples from 85 pneumonia patients were studied. Using CFT, cases were initially classified as due to Chlamydophila (43 cases); to other agents (23 cases) (influenza A and B, Mycoplasma pneumoniae, respiratory syncytial virus, adenovirus and Legionella pneumophila); or as negative (19 cases). Cases were considered positive if they showed seroconversion, a significant rise in titer or high titer; and were finally classified as positive if they gave a positive result in at least three assays. Sensitivity values ranged from 87% to 97.8%; and specificity from 84.6% to 97.4%. In conclusion, the assays compared appear to be useful tools for the diagnosis of pneumonia due to Chlamydophila. Key words: Chlamydophila; Serology; ELISA; Microimmunofluorescence; Complement fixation; IgG. Fernando De Ory, Servicio de Microbiología Diagnóstica, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain. fory/isciii.es Chlamydophila pneumoniae (formerly Chlamydia pneumoniae) (1) is a frequent cause of respiratory infection. Since it is an intracellular parasite, it can only be isolated by inoculation in cell cultures, which requires prolonged incubation periods in order to obtain results. Isolation does not provide a high degree of sensitivity and is not a practical method for diagnosis. In recent years genomic amplification methods have been developed for diagnosing pneumonia due to Chlamydophila; meanwhile, these techniques Received 18 November Accepted 5 January have revealed lack of sensitivity as a method of diagnosis (2, 3) and have not been sufficiently assessed in regular clinical laboratory practice (4). The use of suitable serological tests has led to the recognition of C. pneumoniae as responsible for between 7.5% and 21% of cases of community-acquired pneumonia (5 8). For this reason it is currently recognized that serology is the most suitable diagnostic method. Microfluorescence (MIF), originally designed to detect antibodies against Chlamydia trachomatis (9), provides suitable sensitivity values (10). Amongst currently available techniques of serological diagnosis, the most widely used is MIF. 279

2 DE ORY et al. However, as this assay is technically complex, interpretation of results is subjective, and diagnostic criteria have not yet been harmonized (11), it is only used in highly specialized laboratories. In comparison, the complement fixation test (CFT) offers a practical, low-cost approach, since it permits screening of samples for many different agents within one single assay, though it is time-consuming, difficult to standardize and must be performed by highly qualified personnel with strict quality control. In recent years, a number of commercial assays, including both MIF and enzyme immunoassays (ELISA) for detecting antibodies to C. pneumoniae, have been developed. There is very limited information about the performance characteristics of these assays for diagnostic purposes. The aim of this paper is to compare four commercial ELISAs, one commercial MIF, for detecting specific IgG, and one in-house CFT for total specific antibodies for the diagnosis of C. pneumoniae pneumonia in paired serum samples. MATERIAL AND METHODS One hundred and seventy serum samples (acute and convalescent), taken 2 3 weeks apart, from 85 patients were included in the study. The samples were received in our laboratory (as reference laboratory) for serological diagnosis of pneumonia. No data from direct diagnosis (bacterial or viral isolation, or nucleic acid testing) were available. In 64 cases the age was available (mean 57.6 years, standard deviation 17.6, range 4 months to 94 years). Etiological diagnosis of patients was initially based on either detection of seroconversion (SC) (defined as a change from negative to positive) or a significant rise (defined as a four-fold or higher greater change in the titer), obtained by CFT, against a panel of antigens, including influenza A and B viruses (IAV, IBV), adenovirus (AV), respiratory syncytial virus (RSV), Mycoplasma pneumoniae, and Chlamydophila, or by indirect immunofluorescence for Legionella pneumophila (12). Initially, 43 cases were considered to have been caused by Chlamydophila; 23 cases were due to other respiratory pathogens, including IAV (4 cases), IBV (4 cases), M. pneumoniae (4 cases), RSV (4 cases), AV (4 cases), and L. pneumophila (3 cases). Finally, 19 cases were serologically negative to all the pathogens studied. All samples were tested in parallel using the following methods: 1. Chlamydia pneumoniae-igg-selisa Medac (Medac Diagnostika, Hamburg, Germany) (ELISA- Medac). This method uses a purified C. pneumoniae antigen. The samples were tested at a 1:50 dilution. Cut-off was calculated as the mean of two determinations of a negative controlπ For each sample a ratio was set for the cut-off. Samples were positive if the ratio was Ø1.1. Cases were classified as positive if SC was observed, if the convalescent sample ratio was at least 1.5 times higher than the acute sample ratio, or if acute or convalescent sample gave a ratioø SeroCP Quant IgG (Savyon Diagnostics Ltd., Ashdod, Israel) (ELISA-Savyon). The method uses purified elementary bodies of C. pneumoniae. Samples were tested at a 1:100 dilution. Results were calculated against three calibrators set in binding units (BU)/ml. The BU/ml values obtained in each sample were converted to MIF equivalent end-point titers (EEPT). Thus, results were finally expressed as such. Cases were considered as positive if, first, they showed SC or, secondly, if four-fold or greater changes in EEPTs were observed, or thirdly, if EEPTs were Ø1:256 either in acute or convalescent samples. 3. SERION ELISA classic Chlamydia IgG (Institut Virion Serion, Wurzburg, Germany) (ELISA- Serion). This method uses a genus-specific antigen and samples were tested at 1:20. The results were calculated using SERION easy base 4PL-Software/SERION evaluate-software, by using two determinations of a batch-specific calibrator, and expressed as units/ml. Samples were positive if they gave valuesø15 u/ml. Cases were classified as positive if SC was observed, if the convalescent sample ratio was at least 1.5 times higher than the acute sample ratio, or if a result Ø100 u/ml was observed for either acute or convalescent samples. 4. Chlamydia pneumoniae IgG ELISA (DRG Instruments, Germany) (ELISA-DRG). The method uses the Chlamydophila pneumoniae antigen. The samples were tested at a 1:100 dilution. Results are expressed as enzyme immunounits (EIU), referring to a batch-specific calibrator, tested twice. Cases were classified as positive if SC was observed, if the convalescent sample ratio was at least 1.5 times higher than the acute sample ratio (in samples with EIU values below 130) or 1.3 (in samples with EIU values above 130), or if a result Ø200 EIU was seen in either acute or convalescent samples. 5. MIF, using commercially available slides (Vircell, Granada, Spain), containing C. pneumoniae and C. psitacci and Chlamydia trachomatis antigens in separate wells. Samples were titrated from 1:16 to 1:1024, by using four-fold dilutions. The isotype IgG response was detected by means of an antihuman IgG FITC-conjugated (Dako, Denmark), diluted at 1:400. The reading was done by at least two different people. Cases were classified as positive if, first, SC or a four-fold or higher rise in 280

3 CHLAMYDOPHILA PNEUMONIAE IgG IN PAIRED SERUM SAMPLES TABLE 1. Summary of results obtained by the compared methods Reference Positive Negative (nω46) (nω39) ELISA-Medac Positive (nω45) 42 3 Negative (nω40) 4 36 ELISA-Savyon Positive (nω51) 45 6 Negative (nω34) 1 33 ELISA-Serion Positive (nω43) 40 3 Negative (nω42) 6 36 ELISA-DRG Positive (nω46) 40 6 Negative (nω39) 6 33 MIF-Vircell Positive (nω42) 41 1 Negative (nω43) 5 38 CFT-In-house Positive (nω43) 42 1 Negative (nω42) 4 38 titer was seen or, secondly, if a titerø1:256 was seen in either acute or convalescent sample. 6. CFT followed a standard procedure, as described (13), using commercially available antigens (Institute Virion, Switzerland). Samples were titrated from 1:8, by using two-fold dilutions. All positive cases included in the study showed SC or a significant rise in titer, to the corresponding antigens. All commercial kits were used following the manufacturer s instructions. Samples were stored at ª20 æc until used in the different assays. RESULTS Cases were classified on a consensus basis: a positive result in three or more assays was finally considered to be positive. On the basis of this criterion 46 cases were classified as positive for Chlamydophila and 39 as negative. All six methods showed agreement in 31 positive and 24 negative cases. Table 1 summarizes the results obtained by all the assays, as compared with the reference criterion. Table 2 shows agreement, sensitivity and specificity, and positive and negative predictive values (PPV, NPV) for the methods compared. Agreement ranged from 85.9% (ELISA- DRG) to 94.1% (CFT); sensitivity from 87% (ELISA-Serion and ELISA-DRG) to 97.8% (ELISA-Savyon); and specificity from 84.6% (ELISA-Savyon and ELISA-DRG) to 97.4% (CFT and MIF). The corresponding figures for PPV ranged from 87% (ELISA-DRG) to 97.7% (CFT), and for NPV from 84.6% (ELISA- DRG) to 97.1% (ELISA-Savyon). Tables 3 and 4 show either cases where there was a discrepancy between the results obtained and the reference criterion, or false negatives (Table 3), or false positives (Table 4). Amongst the 19 cases initially characterized as negative for the pathogens studied using CFT, three were finally classified as positive, since either three (cases N14 and N15) or four (N18) positive results were obtained (Table 3). Seven cases showed a single discrepant negative result, two by ELI- SA-Medac (C42 and C43), one by ELISA-Serion (C35), two by ELISA-DRG (C6 and C26), and two by MIF (C31 and C32). All cases with falsenegative results in MIF showed C. pneumoniaespecific IgG but at a very low level (Æ1/64). In cases which were positive for pathogens other than Chlamydophila, one case, positive for AV, was also finally classified as positive for C. pneumoniae as well (AV20, Table 3). In contrast, a single false-positive result was obtained in 10 cases (Table 4): one by ELISA- Medac (IA3), 3 by ELISA-Savyon (N2, N13 and L22), one by ELISA-Serion (N6), 4 by ELISA-DRG (N8, N17, IA2 and RSV4), and one by CFT (C18). The remaining five cases gave a positive result in two assays. TABLE 2.Agreement, sensitivity, specificity and positive and negative predictive values of the compared methods Agreement Sensitivity Specificity PPV NPV ELISA-Medac 91.8% (78/85) 91.3% (42/46) 92.3% (36/39) 93.3% (42/45) 90% (36/40) ELISA-Savyon 91.8% (78/85) 97.8% (45/46) 84.6% (33/39) 88.2% (45/51) 97.1% (33/34) ELISA-Serion 89.4% (76/85) 87% (40/46) 92.3% (36/39) 93% (40/43) 85.7% (36/42) ELISA-DRG 85.9% (73/85) 87% (40/46) 84.6% (33/39) 87% (40/46) 84.6% (33/39) MIF-Vircell 92.9% (79/85) 89.1% (41/46) 97.4% (38/39) 97.6% (41/42) 88.4% (38/43) CFT-In house 94.1% (80/85) 91.3% (42/46) 97.4% (38/39) 97.7% (42/43) 90.5% (38/42) 281

4 DE ORY et al. TABLE 3.Discrepant results (false negatives according to reference criterion) Case ELISA-Medac ELISA-Savyon ELISA-Serion ELISA-DRG MIF-Vircell CFT-In house C6 1.4 Pos 256 Pos 25 Pos 4 Neg 16 Pos 8 Pos C7 1.4 Pos 64 Neg 61 Pos 48 Neg 64 Neg 8 Pos C Pos 128 Pos 50 Neg 136 Neg 256 Pos 8 pos C Neg 64 Pos 3 Pos 19 Neg 16 Pos 8 Pos C Pos 64 Pos 10 Neg 30 Pos 16 Neg 32 Pos C26 2 Pos 128 Pos 10 Pos 149 Neg 64 Pos 8 Pos C Pos 128 Pos 76 Pos 63 Pos 64 Neg 8 Pos C Pos 128 Pos 116 Pos 117 Pos 64 Neg 8 Pos C Pos 256 Pos 19 Neg 118 Pos 16 Pos 8 Pos C Neg 128 Pos 6 Pos 48 Pos 16 Pos 32 Pos C Neg 64 Pos 8 Pos 85 Pos 16 Pos 8 Pos N Neg 64 Pos 35 Neg 17 Pos 16 Pos 8 Neg N Pos 128 Pos 3 Neg 55 Pos 64 Neg 8 Neg N Pos 512 Pos 14 Neg 377 Pos 256 Pos 8 Neg AV Pos 256 Pos 176 Pos 76 Neg 256 Pos 8 Neg DISCUSSION In this assessment the reference criterion used was that of a majority of results from the six compared assays, although MIF is recognized as the best technique for the detection of antibodies against C. pneumoniae (10). Following this criterion, the different assays have, in general, produced acceptable sensitivity and specificity results, which were over 85%. The four ELISA methods studied gave sensitivity values between 87% and 97.8%, and specificity values between 84.6% and 92.3%. These differences do not appear to have been caused by the dilution used in each case, which varies between 1:20 and 1:100. In fact, the assay which analyzes the samples at a lower dilution (1:20) (ELISA-Serion) is not the most sensitive, and shows higher specificity values. Another possible reason for the differences obtained could be the antigen used in each assay. This was ob- tained from C. pneumoniae in three cases (ELISA-Medac, ELISA-Savyon y ELISA- DRG), or was genus specific (ELISA-Serion). Nevertheless, the last assay showed very high specificity, with results lower only than those for MIF and CFT. For this evaluation we have considered in ELISA and MIF as positive the cases showing not only SC but high titer as well. This serological approach has proved useful for the diagnosis of other bacterial diseases such as pertussis (14, 15). In the case of C. pneumoniae infection, however, this approach should be used with caution (10), since background titers to the bacteria are high in the healthy population although they may be present only at low titers. Thus, additional studies on serum samples from the general population are required, especially carried out using ELISA tests. Earlier studies have suggested that CFT sensitivity is very low when compared with other 282

5 CHLAMYDOPHILA PNEUMONIAE IgG IN PAIRED SERUM SAMPLES TABLE 4.Discrepant results (false positives according to reference criterion) Case ELISA-Medac ELISA-Savyon ELISA-Serion ELISA-DRG MIF-Vircell CFT-In house C18 0 Neg 64 Neg 9 Neg 1 Neg 16 Neg 8 Pos N2 1.1 Neg 64 Pos 23 Neg 50 Neg 64 Neg 8 Neg N6 2 Neg 128 Neg 43 Pos 62 Neg 64 Neg 8 Neg N8 2.3 Neg 128 Neg 13 Neg 45 Pos 64 Neg 8 Neg N Neg 128 Neg 7 Pos 328 Pos 64 Neg 8 Neg N Neg 64 Pos 54 Neg 30 Neg 16 Neg 8 Neg N17 0 Neg 64 Neg 3 Neg 396 Pos 64 Neg 8 Neg IA2 0.7 Neg 64 Neg 6 Neg 23 Pos 64 Neg 8 Neg IA3 1.8 Pos 64 Neg 12 Neg 117 Neg 16 Neg 8 Neg IA4 1.8 Neg 256 Pos 22 Neg 205 Pos 64 Neg 8 Neg M9 2.7 Pos 256 Pos 12 Neg 150 Neg 64 Neg 8 Neg RSV Neg 64 Neg 10 Neg 12 Pos 16 Neg 8 Neg AV Neg 256 Pos 31 Pos 40 Neg 64 Neg 8 Neg L21 3 Pos 128 Neg 52 Neg 151 Neg 256 Pos 8 Neg L22 1 Neg 64 Pos 8 Neg 141 Neg 64 Neg 8 Neg diagnostic markers (16). However, in this study it was the technique which showed the highest agreement value. This may in part be due to the fact that the initial selection of cases was done using that technique. Moreover, it is recognized that it is a technique with a certain lack of specificity, mainly due to the cross-reactivity produced by certain enteric bacteria (17). In fact, C18 (Table 4) showed a single positive result using CFT. This did not occur with any other method. The main disadvantage of CFT is the difficulty of standardization, although recently an automated assay was undertaken which proved to be applicable for serological diagnosis of acute respiratory infection from viruses and atypical bacteria, particularly C. pneumoniae (18). PPV ranged from 87% to 97.7%, and for NPV from 84.6% to 97.1%. It is important, meanwhile, to point out that such figures were obtained for a theoretical prevalence about 54% (46 cases from C. pneumoniae infection of a total of 85 cases studies), a figure higher than those obtained in pneumonia prevalence studies (5 8). Some recent studies evaluating the application of commercial kits of ELISA, CFT and MIF have been reported (19, 20). ELISA-Savyon and ELISA-Medac were compared with MIF in their application of C. pneumoniae diagnosis, showing values of sensitivity, specificity, and NPV and PPV similar to those obtained here (20). No previous reports on the performance of ELISA-Serion and ELISA-DRG are available. In conclusion, the assays compared here have proven to be effective for serological diagnosis of pneumonia from Chlamydophila and their use in clinical laboratories will improve knowledge of the real incidence of this bacterium as responsible for community-acquired pneumonia. 283

6 DE ORY et al. This work was supported by grant MPY-1019/04 from the Instituto de Salud Carlos III. The authors wish to thank Isabel Pérez Grajera for excellent technical assistance. REFERENCES 1. Everett KDE, Bush RM, Andersen AA. Emended description of the order Chlamydiales, proposal of Parachlamydiaceae fam. nov. and Simkaniaceae fam. nov., each containing one monotypic genus, revised taxonomy of the family Chlamydiaceae, including a new genus and five new species, and standards for the identification of organisms. Int J Syst Bacteriol 1999;49: Menéndez R, Córdoba J, de la Cuadra P, Cremades MJ, López-Hontagas JL, Salavert M, et al. Value of the polymerase chain reaction assay in noninvasive respiratory samples for diagnosis of community-acquired pneumonia. Am J Respir Crit Care Med 1999;159: Schneeberger PM, Dorigo-Zetsma JW, van der Zee A, van Bon M, van Postal JL. Diagnosis of atypical pathogens in patients hospitalized with community-acquired respiratory infection. Scand J Infect Dis 2004;36: Boman J, Gaydos CA, Quinn TC. Molecular diagnosis of Chlamydia pneumoniae infection. J Clin Microbiol 1999;37: Monno R, De Vito D, Losito G, Sibilio G, Costi A, Fumarola L, et al. Chlamydia pneumoniae in community acquired pneumonia: seven years of experience. J Infect 2000;45: Riquelme R, Torres A, El-Ebiary M, Puig de la Bellacasa J, Estruch R, Mensa J, et al. Community-acquired pneumonia in the elderly. A multivariate analysis of risk and pronostic factors. Am J Respir Crit Care Med 1996;154: Ruiz M, Ewig S, Marcos MA, Martínez JA, Arancibia F, Mensa J, et al. Etiology of communityacquired pneumonia: impact of age, comorbidity and severity. Am J Respir Crit Care Med 1999; 160: Sopena N, Sabria M, Pedro-Botet ML, Manterola JM, Matas L, Domínguez J, et al. Prospective study of community-acquired pneumonia of bacterial etiology in adults. Eur J Clin Microbiol Infect Dis 1999;18: Wang SP, Grayston JT. Micro-immunofluorescence serology of Chlamydia trachomatis. In: de la Maza L, Peterson EM, editors. Medical Virology III. New York: Elsevier Science, Dowell SF, Peeling RW, Boman J, Carlone GM, Fields BS, Guarner J, et al. Standardizing Chlamydia pneumoniae assays: recommendations from the Centers for Diseases Control and Prevention (USA) and the Laboratory Centre for Disease Control (Canada). Clin Infect Dis 2001; 33: Peeling RW, Wang SP, Grayston JT, Blasi F, Boman J, Clad A, et al. Chlamydia pneumoniae serology: interlaboratory variation in microimmunofluorescence assay results. J Infect Dis 2000;181:S de Ory F, Echevarría JM, Pelaz C, Téllez A, Mateo MA, López J. Detection of specific IgM antibody in the investigation of an outbreak of pneumonia due to Legionella pneumophila serogroup 1. Clin Microbiol Infect 2000;6: Grist NR, Bell EJ, Follett EAC, Urquhart GED. Diagnostic Methods in Clinical Virology, 3rd ed, Oxford: Blackwell Scientific Publications, de Melker HE, Versteegh FG, Conyn-Van Spaendonck MAE, Elvers LH, Berbers GA, Van der Zee A, et al. Specificity and sensitivity of high levels of immunoglobulin G antibodies against pertussis toxin in a single serum sample for diagnosis of infection with Bordetella pertussis. J Clin Microbiol 2000;38: de Ory F, Guisasola ME, Téllez A, Sanz JC. Aplicación de un ensayo comercial de ELISA al diagnóstico de la tos ferina sobre muestra única de suero. Enferm Infecc Microbiol Clin 2004;22: Verkkoyen RP, Willemse D, Hiep van Casteren SCAM, Mousavi Joulandan SA, Snijde RJ, van den Bosch JMM, et al. Evaluation of PCR, culture, and serology for diagnosis of Chlamydia pneumoniae respiratory infections. J Clin Microbiol 1998;36: Nurminen M, Leinonen M, Saikku P, Makela P. The genus specific antigen of Chlamydia: resemblance to the lipopolysaccharide of enteric bacteria. Science 1983;220: de Ory F, Guisasola ME, Coccola F, Téllez A, Echevarría JM. Evaluation of an automated complement fixation test (Seramat) for diagnosis of acute respiratory infections caused by viruses and atypical bacteria. Clin Microbiol Infect 2004;10: Persson K, Boman J. Comparison of five serologic tests for diagnosis of acute infection by Chlamydia pneumoniae. Clin Diagn Lab Immunol 2000;7: Hermann C, Gueinzius K, Oehme A, von Aulock S, Straube E, Hartung T. Comparison of quantitative and semiquantitative enzyme-linked immunosorbent assays for immunoglobulin G against Chlamydophila pneumoniae to a microimmunofluorescence test for use with patients with respiratory tract infections. J Clin Microbiol 2004;42: